Colorectal cancer is a leading cause of cancer death in the United States. Many investigators believe that a tumor is derived from a single abnormal progenitor and its descendents. The data supporting this long-held view are neither extensive nor definitive. Moreover, the experimental techniques used were heavily biased. We believe that a tumor is derived from multiple abnormal progenitors because clonal interactions among these cells provide a selective advantage during formation, growth, and progression. This hypothesis will be tested using a unique combination of newly developed mouse models, statistical analyses, and imaging techniques. We will analyze tumors from mice treated with either ethylnitrosourea (ENU) or azoxymethane (AOM), mice in which Apc is inactivated somatically by silencing, and mice in which tumorigenesis is initiated because of a mutation in the TGF2 signaling pathway (Aim 1). If polyclonality provides a selective advantage, heterotypic tumors should be common among these distinct mouse models. The tumors will be maintained in a tissue bank that will be well documented using an Access database. We will explore how heterotypic tumors emerge (Aim 2). Our initial study indicates that the most likely explanation involves clonal interactions occurring over very short distances. We will test whether polyclonality persists as tumors grow and progress from benign to malignant states (Aim 3). The results from our proposed experiments could fundamentally change the understanding of tumorigenesis in the mammalian intestine. The acceptance of this new view will undoubtedly impact the design of approaches for chemoprevention and chemotherapy. Signaling molecules mediating clonal interactions would likely be ideal targets for drug intervention. Potential mediators can be first examined utilizing our tissue bank of highly characterized tumors from a variety of mouse models. Each candidate that is still deemed interesting can then be fully tested using our experimental platform, i.e., one could determine whether elimination of the candidate through either drug intervention or genetic manipulation completely impairs the formation, growth, or progression of polyclonal tumors.